Page 1 of 2 12 LastLast
Results 1 to 10 of 11

Thread: MilkyWay@home News

  1. #1
    AMDave's Avatar
    AMDave is offline Seeker of the exit clause Moderator
    Site Admin
    Join Date
    Jun 2004
    Location
    Deep in a while loop
    Posts
    9,631

    MilkyWay@home News

    New Separation Testing Runs
    Hi Everyone,

    Looking at all of the previous runs (including those from previous projects), there are patterns of interesting and unexpected behavior from the optimization results. In order to better understand these results, I have planned a series of extensive testing runs. These tests will allow us to learn how the optimization routine's responds to specific circumstances and inputs. I will make a comment attached to this post describing the tests & reasons for the tests in greater detail if anyone is interested.

    EVERYTHING IS WORKING AS INTENDED. These results are not entirely understood, not incorrect. I don't expect to invalidate any results from this project, but I do plan on actually producing more, better results as I learn why the optimizer outputs what it does.

    I am rolling out the first 2 series of tests, and have released the runs on the server.

    The names of these runs are:
    de_modfit_14_bundle5_testing_4s3f_1
    de_modfit_14_bundle5_testing_4s3f_2
    de_modfit_14_bundle5_testing_4s3f_3
    de_modfit_14_bundle4_testing_3s4f_1
    de_modfit_14_bundle4_testing_3s4f_2
    de_modfit_14_bundle4_testing_3s4f_3


    Please let me know if you experience any problems with these runs. Thank you all for your help with this project!

    Best,
    Tom

    More...

  2. #2
    Join Date
    Apr 2008
    Location
    NTSC
    Posts
    538
    currently pausing my MW@H and Einstein@Home units on my APU due to crashing/BSoDing... anyone else getting crashing while running GPU projects on APUs?

  3. #3
    Join Date
    Sep 2010
    Location
    Leiden, the Netherlands
    Posts
    4,426
    Remember that m@h needs double precision (though my A10-5700 managed to do m@h on its igp)
    Check if the drivers in use are AMDs adrenalin. Microsoft will replace it with theirs each and every update.
    Last edited by Dirk Broer; 08-06-2019 at 01:17 PM.


  4. #4
    Join Date
    Apr 2008
    Location
    NTSC
    Posts
    538
    yeah, it was still crashing after newest drivers

  5. #5
    Join Date
    Sep 2010
    Location
    Leiden, the Netherlands
    Posts
    4,426
    Couldn't find tasks for your 2400G, whether M@H or Einstein. This presents some difficulties with guessing what the errors were.
    Last edited by Dirk Broer; 08-10-2019 at 02:23 AM.


  6. #6
    Join Date
    Apr 2008
    Location
    NTSC
    Posts
    538
    it's been about a week or a little more since i ran it because it was completely freezing windows (or i guess BSODing after a fair amount of time).

    haven't had freezes since stopping it

  7. #7
    Join Date
    Sep 2010
    Location
    Leiden, the Netherlands
    Posts
    4,426
    How much RAM do you assign to the GPU of the 2400G? I have my APUs on 2GB, if possible in the BIOS -and never less than 1GB.
    And I only run one GPU task at a time on an IGP.
    Last edited by Dirk Broer; 08-12-2019 at 10:21 AM.


  8. #8
    Join Date
    Apr 2008
    Location
    NTSC
    Posts
    538
    i usually run 2 as it nets a notable amount more points (20-30%?) as some bits of the unit seem to fall back to 0% GPU use. i started out at 2GB, but i think i set it to auto as i didn't want some RAM lost due to forced allocation

  9. #9
    Join Date
    Sep 2010
    Location
    Leiden, the Netherlands
    Posts
    4,426
    Moo! can easily run two at a time, MilkyWay and Einstein are far more picky, Seti gives seg fault errrors when running two at a time.
    Since I use Linux kernel 5 all my AM1 APUs are back in business, one even running Moo! with both the AMDGPU and AMDGPU PRO driver.


  10. #10
    Join Date
    Apr 2008
    Location
    NTSC
    Posts
    538
    did something change? i USED to run 2 or 3 at a time with no issues... for several months straight even

    could undervolting have damaged something? (0.1 or 0.09v undervolt to keep it cool enough on the stock cooler until i got a Scythe. however, it's not undervolted now (to keep things simplest))

    p sure it's not the mobo as it's been stable for for the last week or three while i haven't been running it (tho i have been gaming a bit, so it's not like the GPU portion has gone unused).

    no OCing, no PBO. VRM is overkill and good for 150-200 amps: MSI B450-A Pro.

Page 1 of 2 12 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •